Get an inside look at new AI & ML capabilities to help you make smarter decisions. Register for Next '21.

Jump to

Speech-to-Text

Accurately convert speech into text using an API powered by Google’s AI technologies.

  • action/check_circle_24px Created with Sketch.

    Transcribe your content with accurate captions

  • action/check_circle_24px Created with Sketch.

    Deliver better user experience in products through voice commands

  • action/check_circle_24px Created with Sketch.

    Gain insights from customer interactions to improve your service

Benefits

State-of-the-art accuracy

Apply Google’s most advanced deep learning neural network algorithms for automatic speech recognition (ASR).

Easy model customization

Speech-to-Text UI enables experimentation, creation, and management of custom resources.

Flexible deployment

Deploy speech recognition wherever you need, whether in the cloud with the API or on-premises with Speech-to-Text On-Prem.

Demo

Put Speech-to-Text into action

As in this demo, you can easily infuse speech transcription into your applications with the Speech-to-Text API.

Key features

Key features

Speech adaptation

Customize speech recognition to transcribe domain-specific terms and rare words by providing hints and boost your transcription accuracy of specific words or phrases. Automatically convert spoken numbers into addresses, years, currencies, and more using classes.

Domain-specific models

Choose from a selection of trained models for voice control and phone call and video transcription optimized for domain-specific quality requirements. For example, our enhanced phone call model is tuned for audio originated from telephony, such as phone calls recorded at an 8khz sampling rate.

Easily compare quality

Experiment on your speech audio with our easy-to-use user interface. Try different configurations to optimize quality and accuracy.

Speech-to-Text On-Prem

Have full control over your infrastructure and protected speech data while leveraging Google’s speech recognition technology on-premises, right in your own private data centers. Contact sales to get started.

View all features

What’s new

What's new

Sign up for Google Cloud newsletters to receive product updates, event information, special offers, and more.

Documentation

Documentation

Google Cloud Basics
Speech-to-Text basics

Learn the fundamental concepts in Speech-to-Text.

Quickstart
Quickstart: Using the gcloud tool

Send an audio transcription request to Speech-to-Text using the gcloud tool from the command line.

Best Practice
Best practices

Review the best practices for transcribing audio with Speech-to-Text.

Tutorial
ML onramp

Explore Speech-to-Text tutorials, codelabs, and more.

Google Cloud Basics
Supported languages

Learn which languages are available for Speech-to-Text, plus the features and recognition models available for each.

Google Cloud Basics
Speech-to-Text On-Prem

Learn more about Speech-to-Text On-Prem, which enables easy integration of Google speech recognition technology into your on-premises solutions.

Quickstart
Quickstart: Using the Speech-to-Text-UI

Learn how to set up your project and authorization, create and refine a transcription, and learn how to use this configuration in the console.

Use cases

Use cases

Use case
Improve customer service

Empower your customer service system by adding IVR (interactive voice response) and agent conversations to your call centers. Perform analytics on your conversation data to gain more insights into the calls and your customers. Speech-to-Text and its enhanced phone call models are already powering Google Cloud’s powerful solution, Contact Center AI.

Workflow: Data moves from Contact Center Audio Data through Google Cloud products: Cloud Storage to (1) Transcribe with Speech-to-Text API to Natural Language API to (2) Analyze with Cloud Data Loss Prevention to (3) Redact PII with BigQuery. Then flow moves both directions from BigQuery (4) Store to (5) Query and visualize with Visualize Call Data.
Use case
Enable voice control

Implement voice commands such as “turn the volume up,” and voice search such as saying “what is the temperature in Paris?” Combine this with the Text-to-Speech API to deliver voice-enabled experiences in IoT (Internet of Things) applications.

Workflow of voice control using speech to text API. Flow starts with (1) User voice command to User device with two-way flow to (0) Unique secure identity with Cloud IoT Core, and two-way flow to Cloud Functions to (2) Transcribe with Speech-to-Text API to AutoML Natural Language with (3) Intent and entity extraction, back to Cloud Functions and User device.
Use case
Transcribe multimedia content

Transcribe your audio and video to include captions and improve your audience reach and experience. Add subtitles to your content real time to your streaming content. Our video transcription model is ideal for indexing or subtitling video and/or multispeaker content and uses machine learning technology that is similar to video captioning on YouTube.

Transcribe multimedia content workflow: Input Audio Stream through (1) Real-time caption to Google Kubernetes Engine to Speech-to-Text API through (2) Evaluate transcription to Deep Learning VM to Firestore to either Content Moderation Team or through (3) Store and leverage to Broadcast Caption Device.

All features

All features

Global vocabulary Support your global user base with Speech-to-Text’s extensive language support in over 125 languages and variants.
Streaming speech recognition Receive real-time speech recognition results as the API processes the audio input streamed from your application’s microphone or sent from a prerecorded audio file (inline or through Cloud Storage).
Speech adaptation Customize speech recognition to transcribe domain-specific terms and rare words by providing hints and boost your transcription accuracy of specific words or phrases. Automatically convert spoken numbers into addresses, years, currencies, and more using classes.
Speech-to-Text On-Prem Have full control over your infrastructure and protected speech data while leveraging Google’s speech recognition technology on-premises, right in your own private data centers. Contact sales to get started.
Multichannel recognition Speech-to-Text can recognize distinct channels in multichannel situations (e.g., video conference) and annotate the transcripts to preserve the order.
Noise robustness Speech-to-Text can handle noisy audio from many environments without requiring additional noise cancellation.
Domain-specific models Choose from a selection of trained models for voice control and phone call and video transcription optimized for domain-specific quality requirements. For example, our enhanced phone call model is tuned for audio originated from telephony, such as phone calls recorded at an 8khz sampling rate.
Content filtering Profanity filter helps you detect inappropriate or unprofessional content in your audio data and filter out profane words in text results.
Transcription evaluation Upload your own voice data and have it transcribed with no code. Evaluate quality by iterating on your configuration.
Automatic punctuation (beta) Speech-to-Text accurately punctuates transcriptions (e.g., commas, question marks, and periods).
Speaker diarization (beta) Know who said what by receiving automatic predictions about which of the speakers in a conversation spoke each utterance.

Pricing

Pricing

Speech-to-Text is priced per 15 seconds of audio processed after a 60-minute free tier.